Remote Sensing of Riparian and Treeline Woody Vegetation
Combines airborne hyperspectral and lidar data with convolutional neural networks to map and classify shrub and tree species across riparian corridors and alpine treeline zones in western mountain landscapes.
Knowledge Graph (179 nodes, 998 connections)
Research Primer
Background
Riparian zones — the streambed, boggy margins, and uphill water-flow areas surrounding rivers and creeks — are among the most ecologically productive and hydrologically dynamic landscapes in the Gunnison Basin. These narrow ribbons of vegetation and saturated soil regulate water quality, store dissolved organic carbon (DOC), and support disproportionate biodiversity relative to their footprint. Within riparian floodplains, groundwater and surface water mix to create riparian nitrogen cycling hotspots: biogeochemically active zones where enhanced nitrogen transformations (nitrification, denitrification, plant uptake) occur because of fluctuating river stage, bankfull overflow events, and the proximity to stream of organic-rich sediments. Willows (Salix boothii, Salix geyeriana), water birch (Betula occidentalis), dogwood, Baltic rush (Juncus balticus), and creeping spikerush (Eleocharis palustris) anchor these systems and provide the structure that small mammals like Sorex shrews depend upon.
For the Gunnison Basin and western Colorado, riparian function matters because nearly every downstream use — municipal water, irrigated agriculture, fisheries, and recreation — depends on what happens in these corridors. Channel migration, river incision (the cutting down of river channels into bedrock or alluvium over time), and altered sediment grain size all reshape floodplain connectivity. Forced channeling mechanisms, depositional growth on point bars, meander-scale variability, and changes in stream order and stream channel type determine whether floodplains remain wet meadows that filter nitrogen or downcut into disconnected gullies. Quantifying suspended load, shear velocity, stage height, and vertical profiles of sediment and chemistry — often using methods like Wolman pebble counts, sieve analysis, and reduced chi-squared statistics for model fits — is the foundation of modern riparian management.
Historical context
Federal riparian and channel-monitoring policy in the Intermountain West was shaped by a generation of technical guidance produced by the USDA Forest Service, its Rocky Mountain Research Station, and the Intermountain Research Station. The Stream Channel Reference Sites guide (1994) established standardized field protocols for cross-section measurement and longitudinal profiling that land managers across the Bridger-Teton National Forest and beyond still use today Stream Channel Reference Sites. Companion work documenting channel responses to streamflow diversion on small Snake River tributaries demonstrated how water rights and conveyance decisions translate directly into channel form Stream Channel Responses to Streamflow Diversion, while long-term sediment studies on Little Granite Creek established baselines for bedload and suspended-sediment behavior in mountain streams Little Granite Creek.
Wetland and agricultural-impact policy was framed by the 1988 Proceedings of the National Symposium on Protection of Wetlands From Agricultural Impacts, which brought together the U.S. Fish and Wildlife Service, EPA, and Department of the Interior around riparian restoration in places like Muddy Creek, Wyoming and the Green-Colorado system Wetlands Protection Symposium. The USDA NRCS Riparian Assessment (2000) codified how stream incisement and downcutting are evaluated for sustainability across BLM and Forest Service lands NRCS Riparian Assessment.
Management actions and stakeholder roles
Key agencies in the Gunnison Basin riparian-management space include the USDI Bureau of Land Management, the USDA Forest Service (working through the Rocky Mountain Research Station and the Stream Systems Technology Center in Fort Collins), the NRCS, and academic partners such as the Utah Water Research Laboratory. Practical management relies on a toolbox of standardized analyses: WinXSPRO software for channel cross-section and hydraulic analysis WinXSPRO User's Manual, the Erosion Risk Management Tool (ERMiT) for predicting post-disturbance erosion ERMiT User Manual, and photographic utilization guides that allow field staff to assess grazing impacts on key riparian graminoids like sedges and rushes Photographic Utilization Guide.
Management approaches increasingly emphasize process-based restoration. The hydrologic reframing of beaver ponds — showing how they raise water tables, recharge groundwater, and moderate stream temperature — has shifted agencies from viewing beaver as nuisances to recognizing them as low-cost restoration partners Beaver Ponds Hydrology. Floodplain assessments now integrate RTK-GNSS cross-section surveying, dynamic cone penetrometer measurements of soil depth to bedrock, multi-temporal remote sensing channel mapping, and SCREAM (Spatially Continuous Riverbank Erosion and Accretion Measurements) analyses to track channel migration over decades.
Current challenges and future directions
The most pressing issues facing Gunnison Basin riparian systems are climate-driven: earlier snowmelt is reducing peak river stage and bankfull overflow frequency, which limits the floodplain inundation that drives nitrogen cycling hotspots and willow recruitment. Persistent drought, combined with legacy effects from mining overburden and historical grazing, accelerates river incision and disconnects floodplains from their channels NRCS Riparian Assessment. Aeolian dust deposition — monitored through BSNE (Big Springs Number Eight) dust sampler networks — adds another stressor by darkening snowpack and altering runoff timing. Post-fire erosion risk, addressed through tools like ERMiT, is rising as fire seasons lengthen ERMiT.
Future directions emphasize integrating GIS assessments of cottonwood and willow gallery forests, archaeological survey data that documents long-term human occupation of floodplains, and high-resolution channel surveys to identify reaches where restoration — including beaver-assisted approaches Beaver Ponds Hydrology — can rebuild floodplain connectivity. Cloud masking and multispectral remote sensing now allow basin-scale tracking of riparian vegetation condition through time.
Connections to research
Research at the Rocky Mountain Biological Laboratory (RMBL) directly informs and is informed by these management frameworks. RMBL's long-term monitoring of subalpine stream chemistry, snowmelt hydrology, willow communities, and small-mammal populations along the East River and its tributaries provides ground-truth data for the channel-assessment, sediment-transport, and riparian-vegetation protocols developed by the Forest Service and NRCS. Studies of DOC export, nitrogen transformation in floodplain hyporheic zones, and willow demography translate the abstract categories of stream order and channel type into measurable ecological outcomes — making the Gunnison Basin a critical natural laboratory for testing whether national riparian policy actually protects mountain headwater function.
References
A New Hydrologic Perspective of How Beaver Ponds Function. →
A Photographic Utilization Guide for Key Riparian Graminoids. →
Erosion Risk Management Tool (ERMiT) User Manual. →
Proceedings of the National Symposium on Protection of Wetlands From Agricultural Impacts. →
Riparian Assessment, USDA NRCS Bozeman, Montana. →
Stream Channel Reference Sites: An Illustrated Guide to Field Technique. →
Stream Channel Responses to Streamflow Diversion on Small Streams of the Snake River Drainage, Idaho. →
The Nature of Flow and Sediment Movement in Little Granite Creek. →
WinXSPRO, A Channel Cross Section Analyzer, User's Manual, Version 3.0. →
Species (54) →
Juncus
shrubs and trees
Betula
Deschampsia cespitosa
Populus angustifolia
Carex rostrata
Carex utriculata
tufted hairgrass
Calamagrostis canadensis
Salix planifolia
Show 44 more speciess
Eleocharis
riparian plants
beaked sedge
Potamogeton
Salix brachycarpa
Danthonia
salt cedar
Acer negundo
Russian olive
Elaeagnus angustifolia
Salix monticola
Salix bebbiana
Rorippa
Alnus incana
S. geyeriana
Eleocharis palustris
Salix wolfii
S. bebbiana
raptorial birds
Thalictrum alpinum
Salix drummondiana
Agrostis gigantea
Alopecurus aequalis
Salix boothii
Wilson's phalarope
Baltic rush
Juncus balticus
Salix geyeriana
weed species
Muhlenbergia richardsonis
Populus balsamifera
geyer willow
Danthonia intermedia
Salix exigua
Rhus aromatica
Elymus lanceolatus ssp. lanceolatus
Mertensia virdis
Swertia perennis
Phlox longifolia
peachleaf willow
Salix lemmonii
snowberry
Salix candida
S. myrtillifolia
Protocol (2) →
NEON Airborne Observation Platform
High-resolution airborne hyperspectral and lidar data collection and analysis to classify land cover types and derive vegetation parameters for hydrol...
Convolutional neural networks species classification (Plantae)
CNN training using image patches for pixel-based tree and shrub species classification at alpine treeline.
Stakeholder (2)
Intermountain Research Station
USDI Bureau of Land Management
Document (4) →
Riparian Assessment USDA NRCS Bozeman, Montana
Technical report (12/2000). Covers Bozeman, Montana, Missoula. Topics: riparian assessment, sustainability, stream incisement, downcutting. Agencies: ...
Small Mammals: A Beaver Pond Ecosystem and Adjacent Riparian Habitat in Idaho
Wildlife survey (1988-1989). Covers Idaho, Summit Creek, Mackay. Topics: small mammal populations, community organization, beaver pond ecosystem, ripa...
Guide to the Willows of Shoshone National Forest
Walter Fertig, Stuart Markow, U.S Forest Service, Oct-2001
Response of a Depleted Sagebrush Steppe Riparian System to Grazing Control and Woody Plantings
http://www.fs.fed.us/rm/pubs_int/int_rp492.pdf?Warren P. Clary, Nancy L. Shaw, Jonathan G. Dudley, Wictoria A. Saab, John W. Kinney, Lynda C. Smithman...
Dataset (2) →
Sindewald et al - Identifying alpine treeline species using high-resolution WorldView-3 multispectral imagery and convolutional neural networks dataset
The dataset contains region of interest (ROI) polygons for six treeline species found in Rocky Mountain National Park, CO. The tree and shrub species ...
Sindewald et al - Identifying alpine treeline species using high-resolution WorldView-3 multispectral imagery and convolutional neural networks dataset
The dataset contains region of interest (ROI) polygons for six treeline species found in Rocky Mountain National Park, CO. The tree and shrub species ...